Several genome-wide association studies (GWAS) for late-onset Alzheimer's disease (LOAD) have now been published. While all of these studies detected the association of APOE 54 with risk for LOAD only the two largest studies, with over 10,000 cases and controls provided genome-wide significant evidence for any novel loci. To develop larger datasets we and others have formed large collaborative groups, such as the Alzheimer's Disease Genetics Consortium (ADGC). We have also developed an innovative program using cerebrospinal fluid biomarker levels as endophenotypes for our genetic studies of LOAD. Our preliminary data demonstrate the power and novelty of this approach in identifying genes that alter biomarker levels and modify LOAD risk, age at onset or rate of disease progression. This endophenotype approach also has the advantage of pin-pointing specific biological hypotheses regarding the effects of associated variants that can be tested using simple cell culture assays. The goal of this proposal is to combine and analyze existing LOAD GWAS data, then use a novel approach that incorporates quantitative intermediate traits, re-sequencing, bioinformatics, expression and functional studies to facilitate the identification and characterization of genetic variants that modulate risk for LOAD, age at onset or rate of disease progression. To accomplish this we will 1) combine and analyze LOAD GWAS data, 2) use a novel method, the Genomic Information Network, to systematically incorporate biological information to prioritize single nucleotide polymorphisms (SNPs) for follow-up, 3) examine top SNPs from the LOAD GWAS for association with cerebrospinal fluid amyloid-beta and tau levels to establish specific hypotheses of mechanism, 4) use novel genetic and bioinformatic methods to identify putative causal variants from the replicated SNPs, 5) use re-sequencing to identify novel variants in the regions surrounding replicated SNPs, 6) examine the top hits for effects on gene expression. Finally, we will use information from these efforts to test specific amyloid-beta or tau related hypotheses for replicated SNPs in cell culture models. This proposal combines the unique resources and skills of our research team with the vast wealth of publicly available resources into a novel approach to the identification and characterization of genetic risk factors for LOAD. PUBLIC HEALTH RELEVANCE: Alzheimer's disease is a very common neurodegenerative disease with no effective means of prevention or treatment. The goal of the present study is to identify genetic risk factors for Alzheimer's disease, which in the longer term may lead to novel drug targets and improved treatment/prevention of the disease.